Machine Learning

**Machine Learning / Computer Vision Specialist**

Job Description/Posting

**About Us**
Refract Technologies is a family of 60+ members and growing. We are engineers, game producers, artists and entrepreneurs, all with deep passion and experience in the gaming and technology industries. We’re here as veterans from the likes of Electronic Arts, Ubisoft, Unity, IGG, Gumi, XII Braves, Creative
Technologies & more.

Our vision is to extend human experience beyond reality through innovations in gaming and entertainment. We are game changers.

**The Role**

As a **Machine Learning/Computer Vision Specialist ** at **Refract Technologies **, you’ll be part of the team illuminating the future of what image recognition can achieve. This role will see you support the automation of predictive decision-making through the collation, processing and analysis of large populations of data.

If you’re passionate about innovation, and bridging the gap between cutting-edge technologies to mass market consumers, we’d love to speak with you

**Responsibilities**
- Develop and understand image analysis algorithms and SLAM algorithms.
- Build performant mobile based algorithms.
- Create algorithms defining player KPIs and game-event detection in real time.
- Design and create platforms for image processing and visualisation.
- Prepare clear documentation and prepare data for training, retraining and customization of machine learning models internally.
- Draft clear training documents for the machine learning models for external parties to follow.
- Develop Deep Learning models to solve problems.
- Implement full software development: integration, debugging, configuring, testing and deployment.
- Prepare documentation such as technical reports, product manuals, etc.
- Understand dataflow programming.
- Participate in planning, task estimation, requirements gathering, and tracking of work
- Understand the current processes and work within the constraints of the platforms
- Develop quick prototypes to test the viability of the different models.
- Maintain a strong focus on the performance of the model, not just the end-results.

**Traits We’re Looking For**
- Ability to multitask effectively in a dynamic environment.
- Strong time-management and organisation skills for coordinating multiple initiatives, priorities and implementations of new technology and products into very complex projects.
- Strong written and oral communications skills with the ability to effectively collaborate with management and engineering.
- Take a consumer-first based approach to problem-solving.
- Inquisitive and always pushing boundaries with regards to technology to make our products the best they can be.

**Skills and Experience Required**
- Degree in computer vision (CV), machine learning, pattern recognition, or related technical fields.
- Strong mathematical background in Linear Algebra and Computer Analysis.
- Good understanding of SLAM Algorithms.
- Strong skills and understanding of state of the art NN architectures, techniques and principles.
- 3+ years of work-related experience in software development or Machine Learning or high-performance computing.
- Experience working on applied CV projects. Examples include (but are not limited to) optical character recognition (OCR), object detection and recognition, video analytics, image classification, feature/colour spaces, algorithms, filters, 3D Geometric Algebra/Matrice, non-linear optimization, and visual question answering (VQA) etc.
- Good understanding of state-of-the-art deep learning and computer vision techniques.
- Experience in developing supervised and unsupervised learning.
- Experience with CNN, RNN, supervised and unsupervised learning, optimization techniques
- Experience with traditional/legacy computer vision techniques (e.g. OpenCV) and be able to make solid recommendations on techniques based on pros/cons analysis.
- Experience in mobile development and has deployed models onto mobile devices.

**Big Plus**
- Experience with one or more deep learning frameworks such as Caffe, Theano, Torch,

TensorFlow.
- Has prior experience developing or utilising ORB-SLAM.

**Salary**: $5,000.00 - $8,000.00 per month

Schedule:

- Monday to Friday